Search found 2711 matches

by Ben Mathew
Mon Mar 18, 2024 9:09 pm
Forum: Investing - Theory, News & General
Topic: Strange TPAW Result
Replies: 4
Views: 455

Re: Strange TPAW Result

TPAW is a beautiful tool. Thank you very much for creating this! Glad you're liking it! I guess I don't understand something in my first example. Why didn't TPAW calculate a lower spending amount at age 55 to make the portfolio last until age 67, at least at the 50-percentile probability? That would be ideal, but it's hard to calculate. The planner would need to observe the output of the simulation and adjust withdrawals based on that (just like we're doing manually). It wouldn't be an ex ante calculation using a formula. We plan to implement this, but it would be more of a "goal seek" type procedure that's quite different from the rest of the calculations. Would you consider changing the format a little bit to make it slightly e...
by Ben Mathew
Mon Mar 18, 2024 2:58 pm
Forum: Investing - Theory, News & General
Topic: Strange TPAW Result
Replies: 4
Views: 455

Re: Strange TPAW Result

The simulation is showing that with the current plan, you can expect liquidity problems to arise the 50th and lower percentiles. The plan would need to be adjusted to reduce this risk. Here is a plan with inputs from the example above: https://tpawplanner.com/link?params=wXrV9iOrW5jgoIuFUShHzqzQqB0EFy8F Age: 30 years Retirement: 55 years Max age: 100 years Current portfolio balance: $0 Savings: $500 per month during working years Pension: $4,000 per month from age 55 Social Security: $3,000 per month from age 67 As you noted, the spending at the 50th percentile drops down to pension income for about two years before the start of Social Security. At the 5th percentile, it drops for about eight years before the start of Social Security. https...
by Ben Mathew
Fri Mar 15, 2024 12:37 am
Forum: Investing - Theory, News & General
Topic: (Probably stupid) question about online TPAW planner
Replies: 3
Views: 723

Re: (Probably stupid) question about online TPAW planner

gunny2 wrote: Thu Mar 14, 2024 3:38 pm The "Monthly Spending During Retirement" graph...is that what they're saying I'm expected to be able to spend a month?
Yes.
gunny2 wrote: Thu Mar 14, 2024 3:38 pm It would seem so, but their numbers are way high IMO.
These are variable withdrawals that will fully deplete the portfolio—i.e. legacy will be zero.

If you don't want to fully deplete the portfolio, you will need to enter a legacy target under "Adjustments to Spending -> Add Legacy". That will reduce the withdrawals, which might be closer to what you are expecting.
by Ben Mathew
Fri Mar 15, 2024 12:30 am
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

jocdoc wrote: Thu Mar 14, 2024 7:44 am Is there a way to fix the asset allocation, ie not change the risk tolerance with age or is this an antithesis for the raison d'etre (reason) of your planner?
Please ignore this question. I finally figured it out.
Curious to know what you were looking for (and found):

Were you looking to fix the risk tolerance (which can be done in "Risk -> Advanced -> Decrease Risk Tolerance With Age")?

Or were you looking to fix the asset allocation (for which you would need to select the SPAW strategy and then specify a fixed asset allocation)?
by Ben Mathew
Fri Mar 15, 2024 12:13 am
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

I don't understand why extra expense would be a necessary input. I don't have any expenses entered into TPAW and I get what appears to be reasonable savings/AA numbers as an accumulator and withdrawal/AA numbers as a retiree. You don't need expenses in TPAW to know how much you can withdraw to cover gross expenses (including taxes) and deplete the portfolio at end of life. Let's take an example. To keep things simple, let's assume portfolio growth rate = 0% and spending tilt = 0%. Suppose you have a portfolio balance of $1000 and 4 months (Jan, Feb, Mar, Apr) remaining. Each month you can withdraw $1000/4 = $250 each. So your withdrawals will be: Jan: $250 Feb: $250 Mar: $250 Apr: $250 But now suppose that in Feb, you have an extra expense...
by Ben Mathew
Thu Mar 14, 2024 11:52 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

Thanks for describing your processes for obtaining your portfolio balance and for sharing your thoughts on the usefulness of the estimated portfolio balance feature. I will keep these perspectives in mind. We weren't planning on adding new functionalities to the balance estimation until at least duration matching is done, so we'll continue to think about this.
by Ben Mathew
Thu Mar 14, 2024 1:29 am
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

Harry Livermore wrote: Wed Mar 13, 2024 6:53 pm Ah- I think it requires having an "account". I don't have "Go to Main Plan/View All Plans" as a menu item; I only have "Plan From Link"...
Cheers
Yes, plans from links don't have a name. I've added this to the to-do list.
by Ben Mathew
Thu Mar 14, 2024 1:23 am
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

The current portfolio balance is assumed to be VT and BND for the purpose of estimating an updated portfolio balance daily. We will expand the list of funds that can be used for this so you can more closely model your actual portfolio. Note that this does not impact the simulation beyond the estimate for the current portfolio balance. I'd suggest removing this function rather than continuing to develop it. It is no problem to enter in the actual portfolio balance every month and doing so is more accurate unless someone actually has the funds in their portfolio and in the assumed balanced allocation mix, which I bet rarely if ever happens. It is confusing to have this value change automatically each day, especially for a new user. It makes ...
by Ben Mathew
Tue Mar 12, 2024 7:00 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

I'm in the early stages of trying to decide why TPAW needs expense granularity or any expenses at all. If you add one thing, you would eventually need to add everything, then you end up in RPM territory, which is not what TPAW is about in my estimation. Since Allocation is in the title, it would be nice if the model could accommodate any (reasonable) portfolio mix (or more portfolio mixes), possibly drawing on real time (or delayed) CAGRs or other return metric (as it stands, the model is fixed on your two base Vanguard funds; understandable). To me, this would be infinitely more meaningful (and more fun), if not more accurate. The other title denizen is Withdrawal - knowing how much I can pull from the portfolio to cover ALL expenses, no ...
by Ben Mathew
Mon Mar 11, 2024 7:21 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

You can model variable savings in the current planner with a small hack. You just need to say that you are retired and then enter your income during your working years as an income during retirement. My concern with this is the very different tax treatment of income during the working years (in our family's case, I expect income taxes to be considerably higher while we are working than when we retire). I guess one could model that using an essential spending item, or attempt to estimate income net of taxes during the working years. Overall I prefer the far more granular approach of ProjectionLab for our income and expenses during our accumulation years, while finding TPAW extremely helpful for an at-a-glance understanding of potential spen...
by Ben Mathew
Sun Mar 10, 2024 10:10 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

corn18 wrote: Sun Mar 10, 2024 6:55 pm I have every model I could find in my massive retirement spreadsheet. All built in and run automatically. Been retired 3 years and the only output I look at these days is from the TPAW model. Total income + total savings - total expenses. If that is positive, I am happy. Really not much more complicated than that, IMHO.

Although I do take one extra step and multiply the excess income by the VPW constant I calculate for spending between now and age 70 when SS kicks in. Gives me an idea of how much more I can spend each year between now and SS.
Thanks for sharing this. Glad you found a process that works for you.
by Ben Mathew
Sun Mar 10, 2024 5:51 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

Corentin wrote: Sun Mar 10, 2024 3:32 am Wow, fantastic, thank you both ! I expected some food for thought and you served a five-star menu hot on my plate :happy
Glad you like the meal! What's great is that it wasn't hard to prepare. I think that's the mark of a good model. It extends easily and can illuminate many aspects of a problem. The model of supply and demand does that for economics. The lifecycle model does that for investing.

When you push the SWR model and ask more questions, the results quickly become nonsensical and confusing. When you push the lifecycle model and ask more questions, you often get clarity because the underlying framework rests on solid foundations and the analysis extends easily.
by Ben Mathew
Sat Mar 09, 2024 11:20 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

On the other hand, while contributions are typically considered as portfolio-independant - wether they're fixed savings adjusted for inflation in TPAW planner or a percentage of income -, wouldn't it make sense to adjust them as well based on market conditions ? Intuitively, something like : - when the market experiences a downturn and expected returns increase, save more aggressively and delay significant expenses like purchasing a car or a house until the portfolio is closer to its target. - when the market rallies and expected returns decrease, save less and allow for more spending. Is this just naive market timing or is there a way to relate this with the lifecycle investing model in a more formal and actionable way ? I've had the same...
by Ben Mathew
Sat Mar 09, 2024 11:13 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

HOW TO EXTEND VARIABLE SPENDING/SAVING TO THE ACCUMULATION PHASE Hi Ben, I appreciate your understanding of the advantages of amortization versus fixed withdrawals during retirement and the elegant implementation of the lifecycle investing model in your tool. On the other hand, while contributions are typically considered as portfolio-independant - wether they're fixed savings adjusted for inflation in TPAW planner or a percentage of income -, wouldn't it make sense to adjust them as well based on market conditions ? Intuitively, something like : - when the market experiences a downturn and expected returns increase, save more aggressively and delay significant expenses like purchasing a car or a house until the portfolio is closer to its ...
by Ben Mathew
Fri Mar 08, 2024 3:26 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

Interesting, thanks. Extending your example (assuming no change in equity values or expected returns), if yield fell from 3% to 2%, the portfolio bond prices would rise, so the portfolio equity % would decline. The decreased bond market yield would mean a higher ERP, which means a higher Merton share (equity %), triggering a rebalancing and buying more equities, which seems like the right portfolio change in response to lower bond yields. Yes. Note that we would need to rebalance by selling bonds and buying stocks even to get back to the original asset allocation. The increased ERP would mean moving past this original asset allocation target to a new more stock-heavy target, which means rebalancing even more towards stocks. This is the dyn...
by Ben Mathew
Fri Mar 08, 2024 2:12 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

The TPAW Planner says this: “In the case of bonds, it's easier to see why the interest rate for planning purposes should be the interest rate you are currently getting for the bonds in your portfolio, not the interest rates that prevailed historically. For one, changes in interest rates are hard to predict. So the best guess for future interest rates would usually be the current interest rate. Even if interest rates rise in the future, bond prices would fall to compensate. So current interest rates applied to the current portfolio balance may still lead to more accurate estimates, especially if the bonds have been duration matched to reduce or eliminate interest rate risk.” Is my interpretation correct: Let’s say that all my bonds are TIPS...
by Ben Mathew
Wed Mar 06, 2024 6:30 pm
Forum: Investing - Theory, News & General
Topic: Equity Risk Premium when you're not using TIPS
Replies: 3
Views: 778

Re: Equity Risk Premium when you're not using TIPS

If I buy a 20-year TIPS bonds and hold it maturity, my FI portfolio has that real return set ("locked in") for that duration and I shouldn't care what the current TIPS real yield is, right? Yes. What if my "safe" investment for calcuating the ERP is not TIPS but somethign else (for me specifically, it's the TSP G-fund). The no-arbitrage condition would imply that the risk-adjusted yields paid by different bonds are equal. So the real yield of TIPS can be a reasonable estimate of the expected real yield of nominal US bonds, since both bonds are issued by the US government and default risk is negligible. Realized real yields can be pretty different due to unexpected inflation, but not ex ante expected real yields. There c...
by Ben Mathew
Tue Mar 05, 2024 2:05 am
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

PERSONAL FINANCE EDUCATION IN HIGH SCHOOL The article linked in the previous post also talks about a high-school level personal finance class created by Financial Life Cycle Education (FiCycle) : FiCycle focuses more on the underlying concepts [than does the national standard for personal finance education]. It’s about “how and why individuals and households transfer consumption over time,” Financial Life Cycle Education says. From the " Ficycle Standards for Personal Finance and Mathematics ": ... a high-quality education in finance should focus on developing conceptual understanding. Consistent with academic financial theory, the central concept in personal finance is the financial life cycle. Therefore, financial education sho...
by Ben Mathew
Tue Mar 05, 2024 1:43 am
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

PERSONAL FINANCE EDUCATION IN COLLEGE Over the last few years, there has been a push within the economics profession to start teaching personal finance at the college level. This is huge. Financial education has been very lacking. Is there anywhere I can read more about the "push?" One of the economists behind this is Professor Annamaria Lusardi who recently moved to Stanford from George Washington and is Director of the Financial Freedom Initiative there. She posts frequently about personal finance education at the college and other levels. She was also interviewed in the Rational Reminder Podcast . There's a new " Teaching Personal Finance " Conference that she is involved in organizing. The agenda from the first year...
by Ben Mathew
Mon Mar 04, 2024 10:28 am
Forum: Personal Investments
Topic: Retirement Withdrawal Strategy for Preserving Principal and Growing It for Inflation
Replies: 17
Views: 1793

Re: Retirement Withdrawal Strategy for Preserving Principal and Growing It for Inflation

You can model this with TPAW which implements the lifecycle model of asset allocation and withdrawals from economics. The ad hoc SWR model is particularly unsuited to your case because it protects retirement spending (unless the portfolio runs out) and applies most of the risk to legacy. You seem to want the opposite—protect legacy while taking on some risk with retirement spending. The only way to guarantee that your principal is preserved for legacy is to duration match the target using risk free inflation adjusted bonds (TIPS). TIPS goes out to 30 years, so you can plan with certainty how much you can leave to your daughter in 30 years, when you are 90 and your daughter is 43. The real interest rate of 30 year TIPS is currently 2.1%. So ...
by Ben Mathew
Sun Mar 03, 2024 1:06 am
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

If I use the defaults, and set my risk level as 14 ("moderate") on the slider, I end up with an RRA of 2.26, and an AA of 58% equities. OK, fine. If I change the expected returns to "Conservative Estimate" and then go to the risk tab, I see my risk level (still at 14/ "moderate") continues to show an RRA of 2.26. OK. But now the AA shows 33% equities. What gives? If I was planning around a lower expected return for stocks, I think I'd be tempted to do just the opposite- if I believed stocks would return 6%, I might have an AA of 60/40; but if for some reason I believed returns were going to be lower, say 4%, I'd be tempted to change the AA to, say 70/30, because I am reaching for a certain amount of spend and ...
by Ben Mathew
Sat Mar 02, 2024 7:05 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

Ben- reporting my brief history with TPAW and submitting a suggestion for the online planner. I'm an accumulator with just under 6 years to go. 1. Over the last two months, mastered VPW, Pralana Bronze, RPM, and Pralana Gold, in that order 2. Stumbled onto TPAW online (two months ago, I think); my initial reaction was, "meh, kinda basic" 3. I prefer spreadsheets to 'black boxes', so gave OG TPAW a try- hated it; couldn't figure it out 4. Read this thread through and figured out how to modify the spreadsheet 5a. Mastered OG TPAW- still hated it; I think I have a personal vendetta against the layout 5b. I was committed to giving up on TPAW and being a VPW fanboy (while still using Pralana and RPM) 6a. Retry TPAW online (just yester...
by Ben Mathew
Wed Feb 28, 2024 9:38 pm
Forum: Investing - Theory, News & General
Topic: Variable Percentage Withdrawal (VPW) vs. Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 14
Views: 2053

Re: Variable Percentage Withdrawal (VPW)

It sounds like another Monte-Carlo confusing thing, but expected returns in their common acceptance (and as input to any time-value-of-money formula like PMT and NPV) are geometric quantities, not arithmetic. It really beats me why anybody would ever use arithmetic average (mean or median) in ANY context involving returns, to be honest. Returns are geometric quantities by their very nature. I guess old habits die hard. :shock: You must not be familiar with the probabilistic framework used to model risk in finance. Expected return as I've used it is standard usage in academic finance, including for example in Merton's formula. It's the mean of the probability distribution of returns. If you look carefully at the simple example I gave, it sh...
by Ben Mathew
Wed Feb 28, 2024 7:29 pm
Forum: Investing - Theory, News & General
Topic: Variable Percentage Withdrawal (VPW) vs. Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 14
Views: 2053

Re: Variable Percentage Withdrawal (VPW)

For expected returns, enter 6.6% stocks and 2.2% bonds. (This keeps the median return consistent with Longinvest's view of the stock and bond return process. The WAG of 5.0% and 1.9% in VPW is Longinvest's estimate of the median returns for stocks and bonds based on historical data. The expected return will be higher than the median. Hey Ben. I might have asked you already, but I don't remember the answer. VPW's so-called WAG came straight from the historical average return (1900+) documented by Dimson and al, in the Credit Suisse Global Investment Returns Yearbooks. Here is a quick snapshot of the corresponding graph from the 2021 edition. Of course, those values vary a tad from one yearbook edition to another. http://i.imgur.com/hiCTmwFl...
by Ben Mathew
Wed Feb 28, 2024 2:42 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

A discussion about VPW vs TPAW that began in the VPW thread has been moved to its own thread:

Variable Percentage Withdrawal (VPW) vs. Total Portfolio Allocation and Withdrawal (TPAW)
by Ben Mathew
Wed Feb 28, 2024 2:25 pm
Forum: Investing - Theory, News & General
Topic: Variable Percentage Withdrawal (VPW) vs. Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 14
Views: 2053

Re: Variable Percentage Withdrawal (VPW) vs. Total Portfololio Allocation and Withdrawal (TPAW)

Just wanted to let you know I've used TPAW to plan my accumulation for more than a year now. It is easily one of the best resources out there and it's not even close. Like seriously, as someone who leverages, knows and applies lifecycle investing, I understand what it is doing and it is truly unbelievable. Every freaking cell in that spreadsheet is useful. Especially the ones under the savings portfolio heading that basically tell you the AA during each year (and the amount of leverage) is mind-blowingly useful stuff. It's like my own personal finance coach, it figures it all out. If you gave me the option for a free financial advisor for life (any one you want), or this spreadsheet, I'm going with the spreadsheet. That's how game-changing...
by Ben Mathew
Wed Feb 28, 2024 2:06 pm
Forum: Investing - Theory, News & General
Topic: Variable Percentage Withdrawal (VPW) vs. Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 14
Views: 2053

Re: Variable Percentage Withdrawal (VPW)

But my starting position is: Nothing can predict the future While no one can predict the future, all planning methods including VPW and TPAW have to make some assumptions about the future. VPW assumes median returns of 5.0% for stocks and 1.9% for bonds, which translates to expected returns somewhat higher than that. [*] No model can "guarantee success" or avoid failure when faced with extremes All strategies are not equal in this respect. You can structure your strategy to do better when faced with poor outcomes. The obvious lever is reducing risk of the portfolio so it doesn't do as badly when the market does badly. But even for a given level of risk, you can take risk in good and bad ways. To take risk in good ways: Diversify ...
by Ben Mathew
Tue Feb 27, 2024 1:48 pm
Forum: Investing - Theory, News & General
Topic: Variable Percentage Withdrawal (VPW)
Replies: 2256
Views: 605689

Re: Variable Percentage Withdrawal (VPW)

I have no concerns about the assumptions used to build out VPW Retirement model. IMHO anyone with concerns can judge if those concerns are valid by a quick review of the VPW Forward Test. I was already on board with VPW before seeing the forward test, seeing it handle recent market "fun" just confirmed that. I disagree that the VPW Forward Test conveys much information about the long term risks of VPW. Seeing how a plan works over five years does not tell us very much about how safe the plan is over long horizons and poor outcomes. The risk that investors face is just not a bumpy ride with a few temporary crashes. The risk is long term underperformance of the portfolio—specifically the possibility that long term returns come in b...
by Ben Mathew
Tue Feb 27, 2024 11:41 am
Forum: Investing - Theory, News & General
Topic: Variable Percentage Withdrawal (VPW) vs. Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 14
Views: 2053

Variable Percentage Withdrawal (VPW) vs. Total Portfolio Allocation and Withdrawal (TPAW)

[Moved into a new thread from: Variable Percentage Withdrawal (VPW) --admin LadyGeek] My primary issue with TPAW is there's more things you have to figure/guess/enter/etc. I think their argument is that provides more flexibility if you think that VPW's "WAG" (or whatever it's called) at growth rates isn't applicable. OK... But I'm of the view "I don't know what the future holds", and the "less dials I have to play with" the better my outcome will probably be... There aren't more things you have to "figure/guess/enter/etc." in TPAW . TPAW has more adjustments you can make if you wanted to. Most of these are in an Advanced section that you can ignore. The defaults are meant to be a reasonable set of as...
by Ben Mathew
Sun Feb 25, 2024 9:55 pm
Forum: Investing - Theory, News & General
Topic: Asset Allocation - Quantifying Long term Risk / reward
Replies: 31
Views: 4124

Re: Asset Allocation - Quantifying Long term Risk / reward

Thanks again, Ben. Makes sense. I have MIssing BIllionaires arriving today and reviewing the TPAW forum topic has also been very useful. I think my big takeaway from this discussion was that what I was looking for - without initially realizing it - were the models unerpinning lifecycle management. They are, I think, the framework for starting to think systematically about AA questions and understanding long term risk that I started with. Yes, the lifecycle model of economics is where these types of questions about risk, return, asset allocation and withdrawals are explicitly modeled and solved for. The Missing Billionaires is a great place to start learning about the model and how it addresses a wide range of practical financial planning m...
by Ben Mathew
Sat Feb 24, 2024 7:44 pm
Forum: Investing - Theory, News & General
Topic: Asset Allocation - Quantifying Long term Risk / reward
Replies: 31
Views: 4124

Re: Asset Allocation - Quantifying Long term Risk / reward

Edit - I think I found the reason. I had kept the default 100 year max age, had retiremnet at 62 vice 65 and the savings rates were substantially lower in my parameters. When I increase the savings rate post military and drop the max age to 92 the equity allocation recommendations come down. My interpretation - and please correct me if you think something else may be happening - is that these models forced the optimization to have to balance the need for growth to support more withdrawals and compensate for lower savings rate with risk aversion. In effect this need for growth to support future withdrawals increase the utility of growth and effectively "lowers my risk aversion" (or forces me to take greater risk) since I have grea...
by Ben Mathew
Fri Feb 23, 2024 11:00 am
Forum: Investing - Theory, News & General
Topic: Asset Allocation - Quantifying Long term Risk / reward
Replies: 31
Views: 4124

Re: Asset Allocation - Quantifying Long term Risk / reward

What was the asset allocation the planner calculated for you during retirement at the moderate risk tolerance of 12 (RRA = 3.05)? 100% now, 78% at retirement 51 at max age Even with your significant military pension and Social Security, that seems high. You might want to check to see if you've entered any cash flows as an annual number instead of a monthly number. Based on the details you shared in the other thread, I'm getting the following glidepath for a risk tolerance of 12 (RRA = 3.05): 100% now (age 41) 39% at retirement (age 65) 33% at max age (age 95) https://lh3.googleusercontent.com/d/14rnmX96UOyIFJZnHebzlilCC2KxhSHy_ The inputs that generated this glidepath were the following: Age: 41 Retirement age: 65 Max age: 95 Current portf...
by Ben Mathew
Thu Feb 22, 2024 8:47 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

RETURN DATA AND CAPE REGRESSIONS UPDATED Data Update The return data used for the simulations has been updated to the latest available in Professor Shiller's spreadsheet. Previous data: Jan 1871 to Dec 2020 New data: Jan 1871 to Jul 2023 The average historical stock return remains the same at 8.6%. The average historical bond return decreases from 3.1% to 2.8%. (Bonds have done poorly in the last couple years due to the rise in interest rates.) The following spreadsheet shows the updated return data series in the tab "monthly data 1." The old vs new annual return averages are shown in the tab "monthly data_old vs new avg." Excel spreadsheet with the new data: 2024 02 08 shiller download We have made it easier to update ...
by Ben Mathew
Thu Feb 22, 2024 8:26 pm
Forum: Investing - Theory, News & General
Topic: Asset Allocation - Quantifying Long term Risk / reward
Replies: 31
Views: 4124

Re: Asset Allocation - Quantifying Long term Risk / reward

The planner in TPAW does not require the use of surveys to arrive at your relative risk aversion. You simply adjust the risk tolerance slider till you arrive at the spending distribution that you most prefer. So you are discovering your RRA during the planning process in the context of the spending options you actually face. I think that this is a more robust approach than using an out-of-context survey to elicit the RRA. I was playing around with this - the planner is amazing, BTW - but I was a bit surprised to find that under all of the aggressive and moderately aggresive RRAs, the planner had me at 100% stocks through death and I had to go pretty conservative to get it to suggest anything less than 100% stocks now. I assume that what is...
by Ben Mathew
Thu Feb 22, 2024 1:49 pm
Forum: Investing - Theory, News & General
Topic: Asset Allocation - Quantifying Long term Risk / reward
Replies: 31
Views: 4124

Re: Asset Allocation - Quantifying Long term Risk / reward

If I understand correctly - what is happening "under the hood" is we transform portoflio value to utility (via some U(v) function where v is portfolio value) and the expected value of this utility distribution of possible outcomes. I assume that somethign like this is actually meant by "risk adjusted returns". Checking my understanding - If we are truly risk neutral (U(v) is linear) then SD does not matter, we can ignore risk and just look at expected annual return. If (as is the case for me and most people) I have a risk adverse profile, say U(v) = ln(v) (one CRRA/iso-elastic utility that also simplifies math for a toy example), then I transforming the log-normal distribution of portfolio value into a normal distributi...
by Ben Mathew
Wed Feb 21, 2024 1:59 pm
Forum: Investing - Theory, News & General
Topic: Asset Allocation - Quantifying Long term Risk / reward
Replies: 31
Views: 4124

Re: Asset Allocation - Quantifying Long term Risk / reward

100% that the dispersion increases with time. BUT the distribution also becomes much less normal (while I don't know the precise term for this distribution, it has a very heavy right tail). Returns are assumed to be lognormally distributed. i.e. Log returns are assumed to be normally distributed. That makes returns and balances right skewed—i.e. long right tail making the mean higher than the median. Let Balance at time t be X t Return at time t be r t Log return at time is ln(1+r t ) Then X t = X 0 (1+r 1 )....(1+r t ) ln X t = ln X 0 +ln(1+r 1 ))+....+ln(1+r t ) So if log returns are normally distributed, the log balance is the sum of normally distributed returns and so remains normally distributed. But the variance of the log balance in...
by Ben Mathew
Tue Feb 20, 2024 12:31 pm
Forum: Investing - Theory, News & General
Topic: Asset Allocation - Quantifying Long term Risk / reward
Replies: 31
Views: 4124

Re: Asset Allocation - Quantifying Long term Risk / reward

My hunch ... and I would need to do some harder work to prove this unless somebody else has already done it ... is that something like the central limit theorem applies to the net worth of an investment given enough time. The longer the time period (ie. the more random draws for the monte carlo) ... then the tighter the distribution of the possible final net worth. No, this is known as the fallacy of time diversification. See Risk and Time by John Norstad for a detailed explanation. The log final balance is the sum (not the average) of the log annual returns. So the law of large numbers does not apply. The portfolio balance becomes more widely distributed over a longer time horizon. If returns are independent and identically distributed (i...
by Ben Mathew
Mon Feb 19, 2024 5:01 pm
Forum: Investing - Theory, News & General
Topic: Asset Allocation - Quantifying Long term Risk / reward
Replies: 31
Views: 4124

Re: Asset Allocation - Quantifying Long term Risk / reward

You can do this analysis with the online planner in TPAW . Data is US stocks and 10 year Treasury bonds, 1871-2020 The default settings in the planner will (a) change the expected return of stocks and bonds to reflect current yields (b) do a Monte Carlo simulation using 5 year block sampling. To undo this and do a purely historical simulation, you will need to (a) Change the expected returns to "Historical" in Expected Returns and Volatility -> Expected Returns (b) Change simulation method to "Historical Sequence" instead of "Monte Carlo Sequence" in Advanced -> Simulation The problem with the pure historical simulation is that you risk overfitting to past returns. Note that if it really were the case that stoc...
by Ben Mathew
Sun Feb 18, 2024 2:19 am
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

So bonds (ideally duration matched TIPS) exist in both the LMP and RP. The difference is in whether they are rebalanced. The LMP is not part of the rebalancing calculation because the bonds there are supporting the floor. The RP is what is rebalanced to its target asset allocation. For the TIPS in the RP, should the stocks be considered at all when duration matching? I believe there are open debates about measuring stock duration. But if it's estimated and considered, presumably at quite long duration, that would really shorten the target duration for the TIPS. Thank you. In principle the duration of stocks should be considered, but in practice it's hard to. Besides the fact that the duration of stocks is hard to estimate, the expected ret...
by Ben Mathew
Wed Feb 14, 2024 11:37 pm
Forum: Investing - Theory, News & General
Topic: 100% Stocks for the Long Run? Karsten takes on "Beyond the Status Quo"
Replies: 46
Views: 6744

Re: 100% Stocks for the Long Run? Karsten takes on "Beyond the Status Quo"

The arguments for upwards sloping equity allocation glidepaths have never made any sense to me, unless it is specifically addressing the addition of SS later in life, which could justify something like a “bond tent”. What's not to understand? This explains it well: https://www.kitces.com/blog/should-equity-exposure-decrease-in-retirement-or-is-a-rising-equity-glidepath-actually-better/ Note that "the optimal glidepath becomes even more conservative, starting at 10% and rising to only 50%." I understand it. I don’t conceptually agree with it. It seems conceptually flawed although it is hard for me to explain. To put it simply, there is no logic of increasing the risk in your portfolio when the amount of your portfolio decreases an...
by Ben Mathew
Tue Feb 13, 2024 9:03 pm
Forum: Investing - Theory, News & General
Topic: Understanding 4% rule
Replies: 41
Views: 3883

Re: Understanding 4% rule

What am I missing logically ? It's quite simple - you work today for $40k, that is your current income and your current spending is anchored to that, next year your employer decided to cut your salary by 10% and now you only have $36K income, what do you do? adjust your spending needs to your new income or not. That's all there is to portfolio withdrawals, it's an income stream. Whether you are getting a paycheck or that income is coming from your nest egg, they are the same. There is some variability to it. I don't know how people can build a mental image of getting an exact dollar amount income stream for 30 years after working may be 40 years getting used to a variable income stream, which is what we all are used to in our working lifet...
by Ben Mathew
Mon Feb 12, 2024 7:47 pm
Forum: Investing - Theory, News & General
Topic: Understanding 4% rule
Replies: 41
Views: 3883

Re: Understanding 4% rule

swayswift wrote: Mon Feb 12, 2024 6:56 pm What am I missing logically ?
You are not missing anything. This is a consequence of fixed withdrawals that don't not adjust for the portfolio balance. Consider instead variable withdrawals where you adjust withdrawals based on portfolio performance. If you adjust fully to the new portfolio balance and horizon, you won't have the history dependence that the 4% fixed withdrawal methodology has.

You can create variable withdrawals by recalculating SWR withdrawals periodically with a new balance and horizon. Or you can use amortization based withdrawals (ABW) as suggested by the lifecycle model in economics. I compare these two strategies here.
by Ben Mathew
Sun Feb 11, 2024 2:29 pm
Forum: Personal Investments
Topic: Ready to pull the plug - Advice on avoiding SORR
Replies: 29
Views: 2863

Re: Ready to pull the plug - Advice on avoiding SORR

As suggested by the lifecycle model in economics , 1. Amortize the portfolio to calculate withdrawals. See amortization based withdrawals (ABW) 2. Maintain a fixed asset allocation on the portfolio. The asset allocation should ideally adjust for pensions/Social Security. But in your case it's likely small relative to the size of the portfolio, so won't affect the asset allocation much. The above strategy of combining ABW with a fixed allocation minimizes sequence of return risk (SORR) because the same amount of risk is applied to spending at all times. A sequence of poor market returns earlier in retirement is not worse than the same sequence of poor market returns in late retirement. Both result in a decline in spending during the temporar...
by Ben Mathew
Fri Feb 09, 2024 2:31 pm
Forum: Investing - Theory, News & General
Topic: Why can't asset allocations also be passively managed by following the markets?
Replies: 51
Views: 3571

Re: Why can't asset allocations also be passively managed by following the markets?

As exodusing noted above, Bill Sharpe has advocated for holding a global market cap weighted stock-bond fund. But keep in mind that this fund is supposed to serve as the risky asset. Sharpe still mixes it with the risk free asset (TIPS) to arrive at his final portfolio. A global stock-bond fund would not give you your target asset allocation between the risky vs risk-free asset. That's a function of your age and personal risk preferences, and something you would need to determine based on your circumstances.
by Ben Mathew
Wed Feb 07, 2024 5:02 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

ConstantChrysalis wrote: Wed Feb 07, 2024 12:26 pm
Ben Mathew wrote: Wed Feb 07, 2024 11:00 am This is aggressive if there is no spending floor. The labeling of conservative, moderate and aggressive for different RRAs in the risk tolerance slider assumes no spending floor.
Ben, can you clarify what you mean by spending floor? In the planner, there is a Spending Floor setting that is not recommended and there is also Essential Expenses. Which one of these are you referring to?
I meant a funded spending floor, i.e. LMP, that is currently entered through the "Essential Expenses" input. Not the unfunded spending floor entered through the "Spending Ceiling and Floor" input.

I've edited the post to clarify. Thanks for pointing out the potential for confusion.
by Ben Mathew
Wed Feb 07, 2024 4:49 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

Ben, the online planner is amazing! Glad you're liking it! I have a question, if you don't mind. I have created a spreadsheet to play around with different numbers. My numbers in my spreadsheet match your online planner to the dollar on everything except the monthly spending. I assume this is due to the continuous vs discrete compounding. I saw your previous post with the spreadsheet on doing the calculations for continuous to discrete for annual spending. Would you please show your process for doing this on a monthly basis? Both the online planner and the spreadsheet amortize in discrete time. The spreadsheet amortizes at the annual level, and the online planner amortizes at the monthly level. To convert from annual to monthly amortizatio...
by Ben Mathew
Wed Feb 07, 2024 11:00 am
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

Really nice write-up, ConstantChrysalis. I'd like to hear a bit more about Relative Risk Aversion, if you or Ben can speak to that, in the context of Ben's planner. I have just been blindly moving the risk slider until I approximate a 50/50 portfolio (with 2 pensions and 2 social security streams, it hardly feels "aggressive", but that's the label the planner puts this level at) The planner gives me an RRA of 1.44. If I follow your formula: Putting this together gives .032 / (3.05 * .1805^2 ) = I get 68% equities. So, it's the other 4 sources of income acting under the hood in the planner to bring that down to the 50-ish percent? I set a (small) spending floor, and a legacy. Not sure if those have an effect. Sorry, I'm a bit of a...
by Ben Mathew
Tue Feb 06, 2024 7:25 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

ConstantChrysalis wrote: Tue Feb 06, 2024 6:03 pm I hope this comprehensive example of how to use the TPAW Planner with an LMP is helpful.
Thanks for creating and sharing this example!
by Ben Mathew
Tue Feb 06, 2024 7:23 pm
Forum: Investing - Theory, News & General
Topic: Total Portfolio Allocation and Withdrawal (TPAW)
Replies: 682
Views: 166159

Re: Total Portfolio Allocation and Withdrawal (TPAW)

klaus14 wrote: Tue Feb 06, 2024 10:27 am Thank you Ben for creating this awesome tool.
it is the best retirement planning tool i've seen.
it doesn't hide any of the complexity and assumptions one need to make in order to come up with withdrawal numbers.
Thank you for your kind comments!

Financial planning is quite sensitive to the assumptions being made. I think it's important to be able to see under the hood and know what the model and assumptions are. I'm glad you feel that TPAW has done a good job of exposing its parameters.
by Ben Mathew
Tue Feb 06, 2024 2:30 am
Forum: Personal Finance (Not Investing)
Topic: Financial Planner Withdrawal Rate Goes from 3% to 11%
Replies: 24
Views: 2602

Re: Financial Planner Withdrawal Rate Goes from 3% to 11%

Bennie wrote: Tue Feb 06, 2024 2:18 am So this is saying that with the given number of years left (to live) and the given rate of return (now this is monotonic, constant, no real-life dips in some years, right?), *if* you can withdraw (and live on) the given percentages of your total assets, then you will be ok?
Yes, this amortization table (which does not have life expectancy updates) is saying that if your portfolio grows at the expected real rate of return (r) every year till age 100, your real spending will grow by g = 0.75% per year if you follow this schedule of withdrawal rates.